Page 21 - GP Fall 2020
P. 21

24
                                                                                                              !
                                                                                                         23
                                                                                                          !


      A data report was then created for each ex-  value among the groups; IPS e.max Press  opposing enamel compared to the docu-
      perimental specimen indicating the enamel  system  was  39.75  μm ;  IPS  e.max CAD  mented wear values of natural teeth when
                                                              3
      volume and height loss in mu. One investi-  Table II. Descriptive statistics for volume loss (mu)
                                            was  40.58  μm ; Noritake  Super Porcelain  followed up to one year. This one-year wear
                                                        3
      gator performed all the testing.      EX-3 system was 45.08 μm ; and the Lava  value of enamel  opposing monolithic  zir-
                                                                  3
                                            Plus Zirconia system was 48.66 μm .  conia crowns was less than wear produced
                                                                        3
      Statistical Analysis                                                           by popular ceramic materials such as felds-
      A power calculation  was conducted  using        Mean                      pathic porcelain and lithium disilicate. 52
      nQuery Advisor (Version 7.0). Assuming an   Group  N  Volume Loss  SD  Min  Max
      effect size of Δ  = 0.485 (the effect size that   EP  12  39.75  7.33  29  55  Even though previous in-vivo studies have
                  2
      was observed in a pilot study using 3 spec-  EC  12  40.58  13.26  25  72  shown that ceramic materials are wear-re-
      imens per group), a sample size of twelve   SP  12  45.08  16.64  23  82   sistant, 10,16   these materials may damage the
      (n = 12) per group was adequate to obtain a   LPZ  12  48.66  14.85  31  81  opposing enamel, which varies according to
      Type I error rate of 5% and a power greater   E  12  37.08  11.88  16  56  the ceramic material composition. 35-38  How-
      than 99%.                                                                  ever, clinical wear measurements in general
                                            Total  60  42.23     13.38  16  82   are complicated, expensive, and time-con-
                                            Table III. Descriptive statistics for height loss (mu)
      Descriptive statistics (means, SD’s, minima,   Significance  (p-value): 0.225   suming. In addition, such studies can result
      and maxima)  were calculated.   One-way   F-value=1.5, df=4 and 55         in relatively  high standard deviations due
      analysis  of variance  (ANOVA) was used   Levene’s significance: 0.21      to the biological spread between the stud-
      to assess statistical significance.  All anal-                             ied individuals in terms of dietary habits,
      yses were conducted by using the software   Table II. Descriptive statistics for volume loss (mu).  dysfunctional occlusion, biting force, and
      package IBM SPSS Statistics version 19.0                                       bruxism. 21,52      !28
      (SPSS). P-values less than .05 were consid-  Group  N   Mean   SD  Min  Max
                                                        Volume Loss
      ered statistically significant.                                            The results of previous in vitro studies, in
                                             EP    12         33.25  8.2  18  44
                                                                                 which a specific material and the antagonist
      Results                                EC    12          27.91  6.8  18  44  wear of the human enamel were examined,
      Descriptive statistics are shown in Table II   SP  12         34.75  13.2  21  63  have  been inconsistent,  mainly  due to  the
      and Table III, and the data are presented as   LPZ  12          27.50  7.4  17  41  fact that the test parameters differed wide-!28
      side-by-side box-plots in Figures 5, 6.    E  12         29.08  4.6  20  36  ly. 21-25,29,30,34-38,53  Most studies have used flat

      The mean value of opposing enamel volume   Total  60          30.50  8.8  17  63  polished ceramics and prepared enamel
                                                                                 specimens  from extracted  molars as their
      loss for enamel disks of the control group   Significance (p-value): 0.149   antagonists  with  test  chambers  filled  with
      was 37.08 μm , which was the lowest mean   F-value=1.8, df=4 and 55        water and integrated sliding movements in
                  3
                                             Levene’s significance: 0.06         the  wear  generating  processes. However,
                                           Table III. Descriptive statistics for height loss (mu).  there have been huge variations in relation
                                                                                 to force actuators, applied forces, numbers
                                            In terms  of opposing enamel  height  loss,  of cycles, frequencies of cycles per test, and
                                            Lava Plus Zirconia system showed the low-  numbers of specimens. 42,44,49,51,53  Therefore,
 !                                          est mean value of 27.5 μm. The mean value  laboratory  data  cannot  directly  be  verified

                                            for IPS e.max CAD system was 27.91 μm,  with clinical  data, which could also be a
                                            control  enamel  was  29.08  μm,  IPS  e.max  limiting factor in the present study.
                                            Press system was 33.25 μm, and the Nori-
 Figure 6. Box-plots of enamel height loss opposing ceramic and enamel disks’
                                            take Super Porcelain system was 34.75 μm.  The findings of the current study showed no
                                                                                 significant difference between the evaluat-
 specimens.                                 This study showed no statically significant  ed materials and the control enamel group,
      Figure 5. Box-plots of enamel volume loss
                                                                                                                32
      opposing ceramic and enamel disks’ specimens.  differences between the groups in opposing  which is in agreement with Amer et al and
                                                                                          52
                                            enamel volume loss (p-value .225, F-val-  Pathan et al  findings and incompatible with
 !                                          ue=1.5, df=4 and 55) and in opposing enam-  some of the previous studies. 10,11,19,28,31,41-43

                                            el  height  loss (p-value  .149,  F-value=1.8,
                                            df=4 and 55).                        The study also demonstrated  that the use
                                                                                 of monolithic zirconia as a restorative op-
 Figure 6. Box-plots of enamel height loss opposing ceramic and enamel disks’
                                                                                 tion  does  not  cause  significant  enamel
                                            Discussion
                                            The results of the present study demonstrat-  wear and it is within the range of normal
 specimens.                                 ed  no  statistically  significant  differences  enamel,  which is in agreement with other
                                                                                              However, some researchers
                                            between  the  groups in  opposing  enamel  studies.
                                                                                       28,31,32,50
                                            volume and height loss; therefore, the data  reported  less wear  depth  to  human  enam-
                                            supported the null hypothesis.       el with monolithic zirconia compared with
                                                                                 glass ceramic and feldspathic porcelain. 28,31
                                            Data from randomized control clinical trials
      Figure 6. Box-plots of enamel height loss    with a validated method for wear quantifi-  In this study, all ceramic disks were fabricat-
      opposing ceramic and enamel disks’ specimens.  cation are rare, monolithic zirconia crowns  ed following the respective manufacturers’
 !                                          have  shown to  produce  less wear  of the  recommendations  and  then  adjusted  using

                                                                                      www.nysagd.org l Fall 2020 l GP 21
 About the Authors:


 Dr. Roya Zandparsa is a Clinical Professor, Postgraduate Prosthodon>cs and Advanced Educa>on in
 Esthe>c Den>stry, Prosthodon>c Division, TuBs University School of Dental Medicine, Boston, MA.


 Dr. Rabie El huni is an Assistant Lecturer and Clinical Supervisor in the Department of Prosthodon>cs,
 Faculty of Den>stry at the University of Benghazi in Benghazi, Libya.

 !  Dr. Hiroshi Hirayama is a Professor at the Boston University School of Dental Medicine in Boston, MA.


 Mr. Mark Johnson is the president of Texture Technology Corpora>on.
 About the Authors:

 Dr. Roya Zandparsa is a Clinical Professor, Postgraduate Prosthodon>cs and Advanced Educa>on in
 Esthe>c Den>stry, Prosthodon>c Division, TuBs University School of Dental Medicine, Boston, MA.



 Dr. Rabie El huni is an Assistant Lecturer and Clinical Supervisor in the Department of Prosthodon>cs,
 Faculty of Den>stry at the University of Benghazi in Benghazi, Libya.


 Dr. Hiroshi Hirayama is a Professor at the Boston University School of Dental Medicine in Boston, MA.


 Mr. Mark Johnson is the president of Texture Technology Corpora>on.
   16   17   18   19   20   21   22   23   24   25   26